STAT 586 Interpretation of Data (I)

(Modern Applied Statistical Modeling and Computing I)

Spring 2016, Wed 6:40 - 9:30pm
Hill Center, Room 552, Bucsh Campus

Instructor: Min-ge Xie
Office Hours: Wed 1:30pm -2:30pm or by appointment
Office: Hill Center 574, Busch Campus
Email: mxie*AT*

TA: Liwei Wang
Office hours: Tuesday     12:30 - 1:30pm

                    Wednesday 12:30 - 1:30pm
Location: Hill Center 555
Email:  lwwang2010*AT*



Addtional office hours in the final week: To be updated

Prerequisite: Level IV statistics. Recommended 16:960:563, 16:960:582.

Methods of modern data analysis with emphasis on statistical computing. Topics include univariate statistics, data visualization, linear models, generalized linear models (GLM), multivariate analysis and clustering method, tree method, and robust statistics etc. Expect to use statistical software packages, such as R or SAS in data analysis.

Course Syllabus:

         pdf version

Homework Assignments:

         Homework No.1

         Homework No.2

         Homework No.3

         Homework No.4 (optional)

Handout Examples (old files):

         R notes:




        SAS notes:
 and (Box-Cox transformation) (or, note old SAS macro Box-Cox transformation)

Course related notes:

         Regression Overview(Figure1.1;Figure1.1b;Figure1.2a;Figure1.2)

         GLM overview (Figure2.1, Figure2.2, Figure2.3, Figure2.4, Figure2.5, Figure2.6, Figure2.7, Figure2.8 )

Addiortnal Reading materials/notes:

         Bootstrap method (review); Bootstrap Slides

         Extra* GLM notes No1 (Figure3.1, Figure3.2, Figure3.3, Figure2.4)

         Extra* GLM notes No2: (Table1, Table2)  ---

a.        Computer codes (Example5.1 (sas), Example.2 (R/S+),Example2a (sas)), Example3 (sas))

b.       DataSets (Example5.1 Data, Example.2 Data)

c.        SAS micro halfnorm  from M. Friendly's webpage

* These extra GLM notes may cover some topics that are not discussed in class. These notes are for students who like challenges and want to read more materials about GLMs.

Final Project:

 In the final project assignment, you are required to analyze one of your own data sets (if you can identify one, and also please ask me for aproval before you start your analysis) or a data set collected by Dr. Hoover in his class on survey data several years ago  (The data set is protected, so we will provide you the data only in our course Sakai site). Different from the homework assignmentss, you need to (1) form your own questions, (2) try and select statistical methodologies that are appropriate for the particular data, (3) document your analysis and write a report  (including introduction/description of the data, the questions of interest(aims, hypotheses, etc), models/methods used, analysis results, and final conclusions).  The final report is due on May 11 before the final exam, and it should not be longer than five pages (not including computing results/output).


Final Exam:  

The final exam is scheduled  on May 11 from 6:40pm - 9:30pm at Hill Center 552. The exam is cumulative and coves all materials that we learn in the semester.  It is a close book exam, but you can bring with you up to two pages of "cheating sheets". Also please bring with you a calculator and nacessary probability talbles. Laptop computers are not allowed.

  click here for an example problem on the exam

Note: Link to the R software (free)
      Link to Rutgers Site Licenses page